Self-driving cars are capturing news headlines and people’s imaginations. Is it really possible to read a book or watch TV while the self-driving car commutes to work? Is there a need to be in the driver’s seat at all? Confusion reigns for good reason. There are significant capability differences between the assists built into today’s cars and a true self-driving car of the future. But, one thing is certain: THE RACE IS ON.
This AV race has several mountainous challenges: engineering, regulatory, lack of industry standardized technology and tools, consumer trust and acceptance, to name a few. At each progressive level of autonomy the challenges become more difficult. But, among the most mountainous of challenges is data. Underlying an automobile’s autonomous capabilities is volumes and volumes of data, required for both training its AI systems and also for real-time decision making once those same systems are deployed.
It behooves companies to consider data-related processes and infrastructure needs early in research and development to pre-empt the complex issues that arise as operations scale. Without efficient data management, the sheer resources the process will consume can dramatically slow innovation.
Explore the four data considerations in the autonomous vehicle race:
- Data Acquisition
- Data Storage
- Data Management
- Data Labelling
Whether just embarking on aggregating data for AVs or well into the race, companies that are aware and ahead of these data challenges will circumvent issues that could potentially halt their progress.
For those early in the data collection process, consideration of one’s data approach and thoughtful decision-making regarding relevant tradeoffs will help ensure an action plan that is both executable and expeditious. For those where data collection is becoming increasingly precarious, a careful retrofit that leverages what is already in place can take the organization to a more secure, accessible and sustainable data approach.